Counting Vermeer


Appendix IV: Weave Map Scrolls

Each "Scroll" below shows a collection of weave maps. The four large files:

show the weave maps for all the analyzed canvases when calculated at the two different resolutions (0.5 x 0.5 cm square patches and 0.75 x 0.75 cm square patches). The sizes of the map images are directly proportional to the sizes of the canvases so that the images can be easily imported into a graphics program and examined.

In the "allVermeer" scrolls above, each individual weave map (both thread density maps and angle maps) is displayed with a color map suitable to display the full range of values in the individual maps. Thus they are not all directly comparable, since the middle color in one map might correspond to an extreme color in another map.

In order to address this, the bulk of the files in this folder picks a single canvas, which is used to specify the color map for *all* the other canvases in the file. For example, consider the file

vermeerMaps(0.5-2){21, 33, 34, 35}

This uses square patches with 0.5 cm sides to examine L21, L33, L34, and L35. The color map is determined by the first entry in the list, in this case by L21. L21 has two color maps, one for the horizontal dimension (17.51 +/- 1.84 threads/cm) and one for the vertical dimension (14.42 +/- 1.55 threads/cm). The second, third and fourth rows show weave maps for L33, L34, and L35. Each map is presented twice, once with the color map specified by the horizontal map of L21 and once with the color map specified by the vertical map of L21. Thus the leftmost map of L33 (which has horizontal values 13.59 +/- 1.80 threads/cm) appears almost completely black because the color map is centered on the value 17.51. Similarly, the third image in the second row appears very light colored because its mean is 16.56, yet it is displayed in a color map with mean 14.42 (the vertical dimension of L21).

The reasoning behind these choices is that in these scroll maps, possible matches are obvious (because they have the same general appearance). The overly dark and overly light pictures show there is no need to even consider trying to match the weave patterns in those images. For example, it is not necessary to consider possible matches between the horizontal map of L21 and the horizontal map of L33. This is useful for rapidly narrowing down the number of pictures that must be manually studied.

Cookies disclaimer

Our site saves small pieces of text information (cookies) on your device in order to deliver better content and for statistical purposes. You can disable the usage of cookies by changing the settings of your browser. By browsing our website without changing the browser settings you grant us permission to store that information on your device.
I agree